With the rapid increase in longevity and considerable expansion of the share of elderly in the population, it is becoming ever important to understand the mechanisms underlying age-related cognitive declines in order to guide environmental and clinical interventions for older adults and for more accurate prediction of risk for dementia [e.g., Alzheimer?s disease (AD)]. A crucial challenge of aging research is improving understanding of the neurobiological basis of conditions leading to dementia, in order to refine the diagnostic procedures and to target new behavioral and pharmacological interventions. A main avenue for the neurobiological understanding of very early risk processes leading to dementia would be the screening of large populations at risk by means of quick, low-cost, and widely available procedures. The present proposal will test to what extent computerized cognitive tests and portable electroencephalography (EEG) can be used to easily, accurately and efficiently evaluate early cognitive decline in elderly at risk for developing AD. Health disparities represent a critical roadblock mitigating the social and fiscal benefits of early identification and care, not only to the individual patient and family, but also at the state and federal levels. Community-dwelling, African-American elders show faster rates of cognitive decline and are almost twice as likely to develop mild cognitive impairment (MCI) and AD as are older, white Americans. However, they are less likely to be diagnosed or receive treatment in the early stages of these disorders. Development of economically viable and culturally acceptable methods of early detection is critical in minority populations. We propose to identify the first signs of dementia in at risk African Americans with subjective memory complaints (SMC) in their communities within the Detroit metro area by using computerized cognitive tests CogState and NIH Toolbox and a portable EEG. We plan to combine cross-sectional with longitudinal studies: within the first two years, we will cognitively evaluate a total of 500 at risk African Americans with SMC. From these participants we will select 200 who at entry to the longitudinal study will not be diagnosed with either dementia or MCI but will, however, undergo repeated cognitive and EEG testing every 6 months over a 3 year period. We will compare these approaches of repeated cognitive and EEG/ Event-Related Potentials (ERP) testing in terms of their sensitivity of identifying persons already at risk who will progress to MCI and/or AD. Our proposed community based evaluation that combines behavioral and EEG/ERP methods will be used to generate profiles of at risk healthy elderly African-Americans, who may within a short period of time develop MCI or AD. We expect that with the combination of behavioral and EEG/ERP methods, we will be able to develop objective markers, which will reliably identify early signs of cognitive decline. Such markers will be used for identification of significant cognitive decline, and to evaluate efficiency of new pharmacological approaches, which will lead to the development of new therapeutic interventions and disease modification.